A High-Throughput Compute-Efficient POMDP Hide-And-Seek-Engine (HASE) for Multi-Agent Operations

2026-04-29

AI Summary

This work addresses the low sample efficiency and high computational overhead of reinforcement learning in decentralized partially observable Markov decision processes (Dec-POMDPs) by proposing and implementing HASE, a high-performance C++ engine. HASE employs data-oriented design, 64-byte cache-line alignment to eliminate false sharing, and a zero-copy PyTorch memory bridge, enabling efficient execution of mainstream algorithms such as PPO, DQN, and SAC. Experimental results demonstrate that HASE achieves a throughput of 33 million steps per second in single-agent settings—approximately 3,500 times faster than existing baselines—and completes multi-agent policy training within minutes. This substantial improvement significantly enhances the scalability and practical applicability of Dec-POMDPs.

Abstract

Reinforcement Learning (RL) algorithms exhibit high sample complexity, particularly when applied to Decentralized Partially Observable Markov Decision Processes (Dec-POMDPs). As a response, projects such as SampleFactory, EnvPool, Brax, and IsaacLab migrate parallel execution of classic environments such as MuJoCo and Atari into C++ thread pools or the GPU to decrease the computational cost of environment steps. We are interested in optimizing the decision-level of human-AI joint operations, so we introduce a compute-efficient Dec-POMDP engine natively architected in C++ called Hide-And-Seek-Engine. By employing Data-Oriented Design (DOD) principles, explicit 64-byte cache-line alignment to remove false sharing, and a zero-copy PyTorch memory bridge using pinned memory and Direct Memory Access (DMA), our engine sustains throughput of up to 33,000,000 steps per second (SPS) in a single-agent, 1024-environment, decentralized observations on an AMD Ryzen 9950X (16 cores). Ten agents reduces FPS to 7M SPS with generating random actions contributing 1/3rd the total runtime for reference. The engine achieves a throughput increase of approximately 3,500$\times$ over the baseline single threaded vectorized NumPy implementation and successfully trains cooperative multi-agent policies via PPO, DQN, and SAC in minutes, validating both its performance and generality.

Problem

• Dec-POMDP
• sample complexity
• multi-agent reinforcement learning
• compute efficiency
• high-throughput simulation

Innovation

• Dec-POMDP
• Data-Oriented Design
• zero-copy memory bridge
• high-throughput RL
• multi-agent reinforcement learning